Migraine affects 12% of the world population. Chronic migraine, which completely disrupts the lives of sufferers, affects close to 3%, and is extremely difficult to treat. It appears that there is an evolution or 'kindling' from an episodic to a chronc migraine state. Migraine is a pain disorder, but more fundamentally it is a disorder of brain excitability, whose fundamental manifestation is an increase in the 'volume' of sensory input. We have identified changes in brain sensory responses after cortical spreading depression (CSD), a wave of brain activity that is thought to underlie the migraine aura. Remarkably, these changes resemble what is seen in the brain during plasticity - i.e. during learning. Our core hypothesis is that CSD and other migraine-related events co-opt the normal learning machinery of the sensory brain to cause a form of dysfunctional learning - or malignant plasticity. We suspect that the development of an acute migraine attack, and the progression of migraine, is driven by this negative plasticity process. We will use imaging and electrical recordings in mice expressing genes found in migraine patients to test this hypothesis. Our first aim focuses on the changes in individual cellular function caused by CSD, which is thought to underlie the migraine aura. We have identified structural and functional changes in neurons, astrocytes, and blood vessels after CSD, each of which might explain the change in sensory network response we observed. Our second aim broadens our approach to look at larger neural networks disrupted by CSD. In addition to pain, migraine also involves changes in perception of light; photophobia or aversion to light is extremely common. We will image the visual cortex of awake mice as they decide on their preference for light levels, before and after CSD. Our third aim will try to apply the insighs we have gained to the development of migraine treatments. Sensory map sharpening - one of the markers of sensory learning we identified after CSD - may be identifiable in humans using fMRI. We will test whether sharpening of sensory maps occurs on chronic CSD stimulation, and thus might serve as a biomarker of migraine progression. Finally, on the basis of our cellular findings, we will test two clinically-relevant medications whose mechanism may be suppression of excessive calcium activity after CSD. Overall, our experiments should uncover basic mechanisms underlying the development of a migraine attack and the progression of migraine, and point to treatment strategies that arrest the 'malignant learning' that takes place in this disease.